Sentiment classification of Chinese movie reviews in micro-blog based on context

Sentiment analysis at home and abroad has been a hot topic, with the development of sina Weibo and Tencent Weibo and other social networking platform, the micro-blog text sentiment analysis has also been more and more attention. Analysis of the emotional micro-blog text is designed to mining user for a product of positive and negative evaluation, in order to analyze the popularity of products. In this article, we study the sentiment classification of movie reviews in Chinese micro-blog, shows a method of combination of emotional lexicon and Chinese language features. First, we use Douban short commentary as the training data to build a movie field of emotional lexicon, and combination of HowNet sentiment lexicon and the Chinese sentiment polarity lexicon (NTUSD) as the final lexicon. Then, we utilize the characteristics of the Chinese language to define a number of rules to make more accurate on Chinese micro-blog movie reviews sentiment classification. The experimental results show that our method is effective for Chinese micro-blog movie reviews of sentiment classification.